In this paper we introduce a new approach for learning precise and general probabilistic models of code based on decision tree learning. Our approach directly benefits an emerging class of statistical programming tools which leverage probabilistic models of code learned over large codebases (e.g., GitHub) to make predictions about new programs (e.g., code completion, repair, etc).
The key idea is to phrase the problem of learning a probabilistic model of code as learning a decision tree in a domain specific language over abstract syntax trees (called TGen). This allows us to condition the prediction of a program element on a dynamically computed context. Further, our problem formulation enables us to easily instantiate known decision tree learning algorithms such as ID3, but also to obtain new variants we refer to as ID3+ and E13, not previously explored and ones that outperform ID3 in prediction accuracy.
Thu 3 Nov
|15:40 - 16:05|
Friedrich SteimannFernuniversität, Jörg HagemannFernuniversität in Hagen, Bastian UlkeFernuniversität in HagenDOI Media Attached
|16:05 - 16:30|
|16:30 - 16:55|
Shaon BarmanUC Berkeley, Sarah ChasinsUniversity of California, Berkeley, Rastislav BodikUniversity of Washington, USA, Sumit GulwaniMicrosoft ResearchDOI Media Attached
|16:55 - 17:20|
Konstantin WeitzUniversity of Washington, Doug WoosUniversity of Washington, Emina TorlakUniversity of Washington, Michael D. ErnstUniversity of Washington, Arvind KrishnamurthyUniversity of Washington, Zachary TatlockUniversity of WashingtonDOI Media Attached